Research from a Ph.D. dissertation at the Massachusetts Institute of Technology may make it easier for runners to find shoes that fit. It can also help running shoe companies develop better, more innovative products.
Sarah Fay, now a postdoc researcher at MIT’s Sports Lab and the Institute for Data, Systems, and Society, created a model to predict how different 3D-printed shoes will impact different individual runners.
The research was sponsored in part by Adidas, which has an established partnership with 3D printer Carbon. 3D-printing shoe insoles vastly increases the properties that the shoes can have, Fay said, and the way they can be individually tailored.
“The end goal and value of 3D printing is that we can make custom shoes without having to make custom equipment,” she said.
The MIT-developed model could take it a step further, creating technology that could custom-fit shoes based on inputs about an individual runner’s body and gait.
“Ideally, we could make a shoe that's right for you and the way you run,” Fay said.
The model takes into account runner height, weight and other body dimensions, and the properties of the shoes — including stiffness or springiness — and can predict how that individual would run in a particular pair of shoes.
- Sign up for The Beat, BostInno’s free daily innovation newsletter. See past examples here.
The running shoe market is wide, and it’s hard for individual consumers to know how their bodies and running styles will respond to different models. A somewhat recent breakthrough in the competitive long-distance running world was the advent of carbon-plated racing sneakers, which are now the gold standard for elite runners and competitive amateur distance runners alike. The MIT model could predict whether a hobby-jogger would benefit from the shoes, too. With the price tag nearing $300 for this kind of running shoe, that input could be a big deal for consumers.
For running shoe companies, there’s a benefit too, said Fay. Adidas now has the software and the model that Fay developed at MIT, and can use it to test how different runners will react to them without having to go through round after round of trials on runners.
“Our work was trying to be able to predict how the new shoes will perform without having to make a million shoes and put them on different runners,” Fay said.
It offers a way for the company to explore endless different shoe designs without having to spend the money on test shoes and test runners.
A challenge, said Fay, was determining how an individual's running form would change from wearing one pair of shoes to another. The team, including Anette “Peko” Hosoi, professor of mechanical engineering at MIT, included inputs from a publicly available data set that had even more detailed body proportions, to help mimic how different individuals adjust to different surfaces.
Fay’s dissertation has concluded, but there is ongoing work to fine-tune the model and make its applications for individual consumers more accessible, Fay said.
